Project Green-Roof

Estimating roof greenification potential from satellite imagery

Stack:

     
Dr. Markus Meingast ©2019

THE QUESTION

AIRS-Dataset

  • 475km2 of aerial imagery (Auckland, NZ)
  • 7.5cm/px resolution $\rightarrow$ downsampled to 30cm/px
  • 850 images

U-Net Model

  • Encoder / Decoder type neural net (with skip-connections)
  • 39 layers
  • 2.000.000 parameters
  • heavy lifting done on Google Colab (GPU)

Data Pipeline:

  1. AIRS-Dataset: 457km2 of aerial imagery (7.5cm/px resolution)

  2. Image segmentation and preprocessing in Python

  3. Model setup and training on Google Colab

  4. Compilation for Edge-TPU

  5. Live model inferencing through Google Coral TPU Accelerator

In [ ]:
In [ ]:
In [ ]:
In [ ]:
In [ ]:
In [ ]:
In [ ]:
In [ ]:
In [8]:
import pandas as pd
from IPython.display import HTML

import matplotlib.pyplot as mp
mp.style.use(['dark_background'])

import numpy as np
import pandas as pd

import plotly as pl
import plotly.graph_objects as go

import plotly.express as px

import ee

Heading Nr. 2

Some random text

Altair Plots (working with IFrame import)

Out[13]:

Plotly Plot